A fast machine-learning-guided primer design pipeline for selective whole genome amplification

Abstract

Addressing many of the major outstanding questions in the fields of microbial evolution and pathogenesis will require analyses of populations of microbial genomes. Although population genomic studies provide the analytical resolution to investigate evolutionary and mechanistic processes at fine spatial and temporal scales—precisely the scales at which these processes occur—microbial population genomic research is currently hindered by the practicalities of obtaining sufficient quantities of the relatively pure microbial genomic DNA necessary for next-generation sequencing. Here we present swga2.0, an optimized and parallelized pipeline to design selective whole genome amplification (SWGA) primer sets. Unlike previous methods, swga2.0 incorporates active and machine learning methods to evaluate the amplification efficacy of individual primers and primer sets. Additionally, swga2.0 optimizes primer set search and evaluation strategies, including parallelization at each stage of the pipeline, to dramatically decrease program runtime. Here we describe the swga2.0 pipeline, including the empirical data used to identify primer and primer set characteristics, that improve amplification performance. Additionally, we evaluate the novel swga2.0 pipeline by designing primers sets that successfully amplify Prevotella melaninogenica, an important component of the lung microbiome in cystic fibrosis patients, from samples dominated by human DNA.

MultiPrime: A reliable and efficient tool for targeted next‐generation sequencing

“…Conventional PCR primer design tools prove laborious for large‐scale target sets [ 24 , 25 , 26 ], while degenerate primers from sequence alignments [ 27 , 28 , 29 , 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 ] or K‐mer generation [ 40 , 41 , 42 , 43 ] require a delicate balance between coverage and degeneracy [ 31 , 36 , 44 ]. Machine learning algorithms [ 45 , 46 , 47 , 48 ] and other approaches [ 49 , 50 , 51 ] have been explored but often encounter limitations when confronted with the challenge of handling degenerate bases and intricate target sequences. Despite these efforts, the challenge of designing primers for large and diverse targets persists.…”

Targeted Next-Generation Sequencing with MultiPrime A Reliable and Efficient Tool for Global Pathogen Detection

“…However, this method can be time-consuming and labor-intensive when applied to a large-scale set of sequences, making it impractical for high-throughput applications. Second, degenerate primers can be designed using either multiple sequence alignment (Bekaert and Teeling, 2008;Collatz et al, 2022;Fredslund et al, 2005;Gadberry et al, 2005;Hugerth et al, 2014;Jabado et al, 2006;Kreer et al, 2020;Lamprecht et al, 2008;Lane et al, 2015;Linhart and Shamir, 2002;Rose et al, 2003;Yoon and Leitner, 2014;You et al, 2009) or creation of k-mers (Clarke et al, 2017;Hendling et al, 2018;Wu et al, 2020;Dwivedi-Yu et al, 2023) to reduce the number of primers required. However, there is a tradeoff between primer coverage and degeneracy, with high degeneracy potentially leading to amplification of non-target sequences, while low degeneracy may not cover all target sequences.…”

“…To achieve successful amplification of target sequences while avoiding non-target sequences, achieving an optimal balance between coverage and degeneracy is essential (Linhart and Shamir, 2002). Finally, machine learning algorithms (Dwivedi-Yu et al, 2023), genetic algorithm (Huang et al, 2005;Wu et al, 2004;Wu et al, 2009), and others (Haas et al, 1998;Riaz et al, 2011;Smolander et al, 2022) are among the other algorithms that can be employed. However, these programs aim to design primers with perfect matches, and some of these programs only design primers in one orientation while focusing on individual primer coverage rather than primer pairs, potentially resulting in lower coverage.…”

Targeted Next-Generation Sequencing with MultiPrime: A Reliable and Efficient Tool for Virus Detection.

“…The multiple sequence alignment-dependent tools were unable to complete the multiple alignment steps, while the K-mer creation dependent tools failed in the degenerate K-mer creation steps or the following steps. Third, machine learning algorithms (Dwivedi-Yu et al, 2023;Huang et al, 2005;Wu et al, 2004;Wu et al, 2009), and other algorthms (Haas et al, 1998;Riaz et al, 2011;Smolander et al, 2022) can be employed to design primers and then manually create primer sets. These approaches do not include degenerate bases, resulting in low primer coverage.…”